© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 21170
    Prev
    Next
    INDEX
    Explanations

    phrases related to identity declaration and self-reference

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    call or named

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     الم
    -0.25
    {~
    -0.25
    1
    -0.25
     https
    -0.23
    ban
    -0.22
    bnd
    -0.22
    f
    -0.22
    </b>
    -0.22
    SPEC
    -0.22
    https
    -0.21
    POSITIVE LOGITS
     called
    1.38
     Called
    1.36
    Called
    1.25
    called
    1.23
     CALLED
    1.21
     nazy
    1.20
     call
    1.18
    稱為
    1.13
     llamada
    1.11
     noemen
    1.10
    Activations Density 0.067%

    No Known Activations